Skip to main content
Glama

capcut_set_clip

Adjust video and audio properties of a clip at a given start time on a CapCut draft track, such as scale, position, rotation, opacity, speed, volume, fades, effects, and masks. Dry-run by default.

Instructions

Change one clip (the segment at start_s on a track): video scale/position/rotation/opacity, canvas background (color or blur), blend mode, basic adjust, speed (constant), audio volume, fades and noise reduction, stabilization level, audio channel copy, shape mask, flip, crop, curves, color wheels (Pro), HSL (8 colours), voice filter (Pro). Dry-run by default.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
xNoHorizontal position, 1 = half the canvas width to the right.
yNoVertical position, 1 = half the canvas height up.
hslNo{red|orange|yellow|green|cyan|blue|purple|magenta: {hue, saturation, lightness}} CapCut UI -100..100; {red: null} removes that colour.
lutNo{path: absolute .cube/.3dl, intensity: 0-100 (default 100)}; null removes the LUT.
cropNo{left, top, right, bottom}: fraction cut from each side (0..1); null resets to the full frame.
flipNo{horizontal?: bool, vertical?: bool} (mirror).
maskNo{shape: line|mirror|circle|rectangle|star|heart, x, y (offset / half canvas, -1..1), width, height (fraction of canvas; defaults 0.28, 0.5), rotation (deg), feather, round_corner (UI 0-100), invert}. null removes.
alphaNo
draftYesDraft folder name inside the drafts root (from capcut_list_drafts).
scaleNo1 = fit (100%).
speedNoConstant speed; the timeline length becomes source length / speed.
trackYesTrack name (main, music, ...).
adjustNoVideo basic adjust in CapCut UI values: exposure, contrast, saturation, temperature, tint, highlights, shadows, whites, blacks, brilliance, vignette (-50..50); sharpen, clarity, grain, fade (0..100). 0 removes the slider.
curvesNo{luma?|red?|green?|blue?: [[x, y], ...]} 2-16 points in 0..1, x ascending, e.g. luma [[0,0],[0.58,0.73],[1,1]]; null removes.
volumeNoLinear gain: 1 = 0 dB, 0.1 = -20 dB, 10 = +20 dB (CapCut max).
denoiseNoCapCut 'Reducir el ruido' (local model; Pro in CapCut 9.5.0). false removes it.
dry_runNotrue (default): return the plan only. false: write to disk (CapCut must be closed).
start_sYesStart of the segment to change.
rotationNoDegrees, clockwise.
fade_in_sNoAudio fade-in (audio clips and the sound of video clips).
stabilizeNoVideo stabilization: 0 off, 1 minimum crop, 2 recommended, 3 most stable (Pro in CapCut 9.5).
volume_dbNoSame in dB as the CapCut volume panel; -60 or lower = mute.
backgroundNo{type:"color", color:"#RRGGBB"} or {type:"blur", blur:0..1 (CapCut levels 0.0625, 0.375, 0.75, 1)}.
blend_modeNo
fade_out_sNo
color_wheelsNoPro. {lift?|gamma?|gain?|offset?: {red, green, blue, luma, saturation} raw -1..1, intensity 0..1}; null removes.
denoise_rateNo
voice_filterNoCapCut voice filter (Filtros de voz, Pro in 9.5.0): {resource_id, name?, strength 0..1}, e.g. 7320193577613529602 = Dulce; null removes.
audio_channelNoCopy one audio channel onto the other (CapCut 'Completar canal').

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.3.0

TDQS

A3.9/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the burden and does disclose the key trait 'Dry-run by default', plus Pro-gated features (curves, color wheels, voice filter). It does not spell out what happens to unset properties or failure modes, but the dry-run default is meaningful safety context for a mutation tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The verb+resource and scoping constraint are front-loaded, and the capability list, though long, is justified for a 29-parameter tool. It is a single dense run-on sentence, so a slightly more structured list would read better, but nothing is wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 29-parameter mutation tool with 86% schema coverage and no output schema, the description covers the capability surface and the dry-run default adequately. It does not explain return behavior for dry-run, but the schema's dry_run text carries that, so the gap is minor.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 86%, so the schema already documents parameters thoroughly. The description's enumeration (scale/position/rotation/opacity, HSL, curves, etc.) maps to params but adds little syntax or meaning beyond what the schema provides; '(constant)' and '(8 colours)' are minor additions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb (Change) and resource (one clip, the segment at start_s on a track) and then enumerates exactly which clip properties it governs. This distinguishes it cleanly from siblings like capcut_set_speed_curve, capcut_add_filter, and capcut_style_text, which handle other aspects.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage is only implied: the agent infers this is the tool for editing an existing clip's properties. There is no explicit when-to-use, when-not, or routing to alternatives such as capcut_set_speed_curve for variable speed (the '(constant)' hint is a nudge but not guidance), nor any stated prerequisites beyond what the schema carries.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.